NefMoto

Can someone confirm if the variables:pvdk_w, pus_w, fvpdkdsl_w, fho_w, frhodkr_w, and SDLDSUA axis calculation have been changed correctly using IDA in this bin?? The rest of the BIN is bone stock so ALL of the maps will need scaled as in the outline, but if the ASM stuff has been done correctly in IDA this will be an excellent starting point for the 1.8T guys (myself included) as I have verified 032PL can be crossflashed to 032HS without immo off.

Thank you

Justin

IMO this is worth a community thread.

Split accordingly.

If i get a chance I will take a peek at the image

Thanks Nyet.

Went through the bin and scaled all of the maps listed on the wiki that were defined in the XDF's I have. Here is the scaling in progress BIN with a txt document outlining what was done, found, not found etc.

Thank you for taking the initiative. It is appreciated and valued.

Sadly, I may not have a chance to look at your bin since it would require me also getting IDA up and set up with that BIN

Will take a look at this as soon as I can find time.

I have a 8N0906018AH (AMK engine) software with all the asm patches and map pack with all the maps that needs rescaling marked. I will share this too.


Sent from my iPhone using Tapatalk

Everything is in the right place, except the SDLDSUA axis.

In ME7.5 we have KFSDLDSUA and its axis is not hard-coded. Its a 16x6 map, located at 15B32 in your file. X-axis, witch is pressure, is located at 15B12. Y-axis is at 15B0C.

Thanks for taking a look Noice, but I opened the bin and XDF and confirmed the x y axis and map location (KFSDLDSUA) are correctly defined in the XDF???

I didn't opened the XDF to check.

The address I gave are based on disassembling and they are correct.

gotcha, sorry I thought you were saying they were incorrect

Hopefully can keep this project going. I think I have found the extras that you were missing.

0x1498E;"Integratorbeiwert für Saugrohrmodell (Dynamik)";"KISRM";"1x1"
0x27D3C;"Konstante für Regeneriermasse in Leitung TEV-Saugrohr bei Normdruck + Normtemp.";"KMLTESG";"1x1"
0x1C83E;"Ersatzwert für Umgebungsdruck";"PUE";"1x1"
0x1C816;"Delta Druckwert für Signalvergleich Ladedruck mit Umgebungsdruck";"DPDSVLU";"1x1"
0x1C840;"Umgebungsdruckbegrenzung minimal";"PUMN";"1x1"
0x1C842;"Umgebungsdruckbegrenzung maximal";"PUMX";"1x1"
0x1C846;"pvdk min. Schwelle für Fehlererkennung Ladedrucksensor";"PVDKMN";"1x1"
0x18A25;"Zeitkonstante für Filterung von pu und fho in GGDSU";"ZDSU";"1x1"
0x28A5A;"Ersatzwert für Umgebungsdruck für BKV";"PUEBKV";"1x1"
0x15B32;"Schwelle für dynamische Schubumluftventilsteuerung aktiv";"KFSDLDSUA";"16x6"


Uploaded KP. I was going to add them to the XDF but didnt want to step on any toes. I cant contribute much more than this but would love to learn about the ASM changes as I'm a noob with IDA.

If you want to add them to the XDF there is no stepping on toes as far as I’m concerned. This is for all of the 1.8t community and I appreciate your help.

What's the point in doing this on some obscure bin?

It's enough to do one bin for every bootrom and then crossflash ECU to it.
And I'm pretty sure there's a bin that's already done.

Audi A4 and A3 have same bootroom. But things like ESP, transmision maybe can't be coded after croshflashing.

Updated the XDF with what I believe are the extra ones you were missing.

I'm unsure of ZDSU which is currently 3 seconds (supposed to sqrt and multiply by two gives 3.46, will only accept a value of 3 or 4) does this need to be changed or can it stay original?

Another i'm unsure of is KMLTESG - value in table is 0.00001600 but wont allow for it to be divided by two, it'll only accept zero or the original, can't go lower.

Hopefully someone more experienced may be able to clear this up.


PRJ - Would something like 06A906032HJ be a better option for cross compatibility? personally I don't mind which binary, i'd just like to learn how to do it so I can do the binary i'm using once i have the know how!

I could not find one that looked right.

I use 8N0906018J on these, because there is a full hex/a2l available.
This is not an A4 file, it's A3/Golf/Leon/Octavia platform.

And if you are doing 5120, start with a file you have hex/a2l for at the VERY LEAST.

Most up-to-date XDF

Either tonight or tomorrow I will post the BIN, XDF and log.....but we’ve got PID control at 26 psi!!!  8) 8) Thank you to everyone that has helped!!

BIN, log and XDF I used. The boost pressure isn't scaled correctly with visualme7logger or vag com so I'll have to look into that, but it functioned great in the car and I had 26 psi on the boost gauge during the log.

this may help

http://files.s4wiki.com/ME7L-ecu/06A906032LP-typical-5120.cfg

I also have a tool that replaces all the scaling automatically in perl


https://github.com/nyetwurk/ME7L/blob/master/logs/double-mbar.pl

Thank you Nyet. I was trying to find that first doc last nite. I knew I had run across it somewhere.

I actually posted the wrong one

http://files.s4wiki.com/ME7L-ecu/my_06A906032LP-5120.ecu

Got the logs showing the correct pressure. Thank Nyet. On a sad note looks like I'm out of turbo  :(

http://nefariousmotorsports.com/forum/index.php?topic=3027.msg105771#msg105771

.ecu file that has been modified.

Also, not that most of you don’t already know this, but it should be said that the bin I posted is absolutely NOT stock rod safe. Torque is coming in as early and aggressively as the turbo will allow. I broke stock rods in another car of mine with a MUCH mellower torque curve. I am also running E85. I think the knock sensors would lose their minds on premium with the timing/boost in the bin. Just saying.

No need to say anything, those who flash random tunes will just flash random tunes.
You're not liable for that in any way.

If you got it running, it's nice. I would always start with a file you have a full damos for though, otherwise there's a lot of extra work involved in searching for parameters, which there doesn't need to be.
There are a2l/dam files available for every type of 1.8T configuration that exists, and then it's just a crossflash away.

Thanks prj. And yes it's working well. I actually just started modifying boost (torque) by gear. It's nice to be able to mat the gas pedal in first gear without just blowing the front tires off the car.

Sorry to slightly divert the thread, I've been attempting to complete the 5120 hack on my own binary. I successfully loaded the rom and flash into IDA (Thanks to PRJ for his awesome autoIT script, and Nyet's mod to the script to suit my funky torrent version of IDA)

I managed to find the parts I need to divide and manually adjusted them in IDA. I finished up yesterday evening but now i have no idea how i export the newly changed binary from IDA?

Probably a very nooby question but i'd appreciate some tips on how to do that. Thanks in advance guys.


If i can get it to work i'll post some screenshots to maybe help some other noobs complete the hack.

You know offsets you know values
Just change it in flash with any hex editor

I think I'm out of turbo  :( Also, and I know fitting a physical EGT sensor is the only way to know for sure, but does anyone know if calculated EGT is artificially higher than actual running E85 as there is obviously no way for the ECM to know what's getting burned in there.

I've been working on this recently for my own binary and believe I've made some progress so I'll document it as best I can to hopefully help some others.

I used the AutoIT script PRJ created to make it easier to load my binary into IDA to find the locations of the ASM divisions. To do this you need to split the flash 1024kb into a ROM and Flash section. The ROM section is the first 64kb of the file you'd read from the car. I used a hex editor to do this and then ended up with

ROM 64KB (0 to FFFF)
FLASH 960KB (FFFF onwards to FFFFF)

It's necessary to do this to allow the loading of the binary into IDA.

Once loaded into IDA I needed to find the locations of the constants that need to be divided.

There are

3 x 'mov r3/4, #654Dh' to find and 5 x 'mov r2, #287h'  (not sure if these are commands, functions or routines) I don't know the correct terminology so please enlighten me lol

I believe the SDLDSUA axis division is not required in ASM for my binary but instead I am able to divide the KFSDLDSUA axis in the table

I didn't have the skills to make the necessary divisions within IDA as I'm not sure how to do that, so I noted down all the addresses of the divisions I needed to make.

I found what I needed at the addresses below (your binary will be different) now this is where I may have messed up, if I have not already by this point. I then opened the binary in a hex editor and attempted to find what I needed. I set it to display 16bit and went to the addresses and was unable to find 654D or 287 values in the binary at those locations, until I added the ROM size to the addresses I found.

Please see below for the offsetting of the addresses


mov r3/4 #654Dh

0x52B48            WITH ROM FFFF ADDED  0x6B248
0x52D1E           WITH ROM FFFF ADDED  0x62D1E
0x556B2            WITH ROM FFFF ADDED  0x656B2

mov r2, #287h

0x54510            WITH ROM FFFF ADDED  0x64510
0x53FAC           WITH ROM FFFF ADDED 0x63FAC
0x53DE8           WITH ROM FFFF ADDED 0x63DE8
0x53D5A           WITH ROM FFFF ADDED 0x63D5A
0x53D0C           WITH ROM FFFF ADDED 0x63D0C


Once found in the binary I divided them in the binary.

654Dh became 32A6h
287 became 50E


Hopefully someone more experienced can chime in and tell me if what I've done sounds correct or not.

I have a couple of queries regarding the following that someone may be able to advise on

- KMLTESG-Constant for regeneration mass in line TEV suction pipe at standard pressure + standard temp

 (stock value is 0.000016, I need to divide this but winols only allows 0 or the original number) am I ok to leave that as it is, or will it cause issues?

- ZDSU

Value has to be sqrt and multiplied by 2. Stock value in my binary is 3, sum gives 3.464 if im correct. I'm only able to change it by whole numbers it seems, so its either 3 or 4. Again not sure if this will be fine left as is or not?




Finally – I've seen warnings from the likes of Bische in the original 5120 threads that the brake force amplifier needs to be disabled via CWGGPBKV, in some binaries it is already disabled with a value of 6, my stock value is 2 so i'm not sure if chaning CWGGPBKV is enough.

Comments, criticism and advice are appreciated.

 KMLTESG-Constant for regeneration mass in line TEV suction pipe at standard pressure + standard temp

 (stock value is 0.000016, I need to divide this but winols only allows 0 or the original number) am I ok to leave that as it is, or will it cause issues?

- ZDSU

Value has to be sqrt and multiplied by 2. Stock value in my binary is 3, sum gives 3.464 if im correct.

In my bin I left KMLTESG at 0.000016 and ZDSU also at stock value of 3. I’ve gone through three tanks of fuel 5120 and haven’t experienced any problems.

The only thing I have noticed and been unable to come up with a logical explanation for is only seeing 195-200 g/s when logging MAF. Fueling looks good in the logs and the car rips, so it almost has to be an improperly scaled MAF, but i followed the wiki, did the 200 offset thing, etc. Any ideas how my MAF can be reading that low on a car that spins the tires hard on a 30 mph 2nd gear pull?? Like damn near smokes the tires spinning.

Lambda control could be maxed out, or very near at WOT which would indicate maf reading a bit low, maybe worth a check.


Just need to figure out what I need to do with the brake force amplifier now - I heard this affects my binary and a few others like 032HJ, HS, TL, QR, RP.

Thank you I will. Also I was thinking about it. Is brake force amplifier the brake vacuum pump?? The automatic transmission (vws) were equipped with an electric pump, the manuals were not. Maybe??

I also (at the end of my log pulls) have hit the brakes HARD at 95+mph to get back down to a sane speed ASAP many times since 5120 and have not noticed any difference in braking response from before I went 5120. For what that's worth.

is ps_w close to actual boost? If so, your MAF is scaled fine.

I'm still unclear on how this is manifest. AFAIK it doesn't exist on 2.7t 6sp ME7.1 so I am totally clueless.

I’ll try have a look at the FR and understand what it does/is doing. It was just something I noted down when reading the 5120 hack stuff. I just need to order a sensor now and then I can test it.

I posted regarding the ASM changes and hope I have them correct!

I will log ps_w tomorrow and post the log. Thanks Nyet

I've never heard about that brake amplifier thing either. As for hardware on 1.8/T b5/b6 manual it is full mechanical vacuum "booster" no electronics in it.

Maybe I've totally thought up a none existent problem then lol my brake booster is all manual also no electronics directly connected to it

http://nefariousmotorsports.com/forum/index.php?topic=2747.75

Bische mentioned it here

WARNING!

Before attempting any pressure scaling you must check if your car uses brake force amplifier(%GGPBKV)!

I just became aware of this, and if this function is in use it must be scaled. I dont really know what would happen if not scaled, but I dont want anyone to have their brakes messed up!

In the 1.8t and S4 2.7 binarys I have with this bit defined, it is CWGGPBKV = 6dec. 6dec = Brake amplifier function turned off.

Please verify your CWGGPBKV in %GGPBKV if you are going to scale the file.



These are some binaries that have it set to 2 stock

O32RP, 032QN, 032TL, 032HN CWGGPBKV = 2.

Well if someone wants you could track down in FR to see what it does exactly because either its an auto thing or idk. Since the vacuum booster only works when you are in vacuum( which is okay since you are never braking with gas on) I dont see what it has to do with pressure values, and you cannot left foot brake with me7 in stock form anyway.

The electric brake vacuum pumps on the automatics could be activated via output diagnostic test mode in the ecm. That said I don't think any of the jetta's, golfs, gti's or beetles had any kind of a pressure (vacuum) sensor in the booster line that was monitored by the ecm. The Passat's and touaregs did though. So my assumption is the sensor would have been common on the audi's, but my audi specific knowledge is limited as I don't work on many of them in a vw shop.

Electrical brake booster in Audi exist only with automatic gearboxes and after october 2000 , there is a pressure sensor on the vacuum line between the firewall and brake servo. (https://uploads.tapatalk-cdn.com/20200629/c47ca791841c18ace4f614adb1701ef1.jpg)

More info is in ssp https://www.google.com/url?sa=t&source=web&rct=j&url=http://www.volkspage.net/technik/ssp/ssp/SSP_257.pdf&ved=2ahUKEwi12YOQsKbqAhXrkYsKHU7AADAQFjAAegQIARAB&usg=AOvVaw2bzYoTDcU-IgRfpgMVqp1n

::thumbs up::

Finally got a log with ps_w. I don't know how dead on it should be if the MAF were scaled perfectly, but it looks pretty close through the log to me. Or is that not close enough and I should touch up the different areas of MLFHFM.

You're good, its fine.

A bit of underscaling never hurt anyone as long as your fueling is good.

Excellent, thank you. Just wanted to make sure I was interpreting what I was seeing correctly.

Hi,

did not want to start new topic, but I would like to hear your opinion about my log...
I have not had much time to play around with that 5120 hack, but since I got the sensor
back on and I want to get it working as it should.

1.8t, gt2871r, 730cc, 3,25" maf

I want to try 1.7bar of boost, but something is limiting/cuting it.

Actual boost vs requested are not maching up...
Is my problem caused because of KFMIRL/KFMIOP/LDRNX not set up right (asking not enough load)?
or is it something else... can someone point me to the right direction please.

There are 2nd and 3rd gear pull in the log. 2nd looks fine, but 3rd spikes badly...

?

So you think that kfldimx is way off? I have scaled the axis and I know I should change the values in it... as I did today... But I'm not sure is it the main problem

Also this was 2th gear pull... 3th gear goes way over the requested load

I have noticed on my car: 2nd gear pull 22 psi, 3rd gear pull 26 psi, 4th gear pull 28 psi. I’m assuming this is due to the load change and using LDRXN for boost request with KFLDHBN up and out of the way. Between busy life and work I haven’t had time to investigate any further. At some point I am going to change my strategy and get LDRXN out of the way and use KFLDHBN to control boost as I suspect this will correct my boost pressure changing based on gear/load problem

most up to date E85 .bin

3 bar map sensor?
With your old version file i received "The 17705 Pressure drop between turbo and Throttle Valve (check D.V.)" with a 4 bar sensor in VCDS.
Something is wrong in file?

I have put several thousand dtc free miles on my car with the bin as posted. There is nothing wrong with the file for MY car. The idea behind posting the file publicly was to make the ASM stuff accessible for the 1.8T vw guys and to document the changes necessary to run E85 in my car, with my turbo, my injectors and my fuel pump. It is posted as a tool to learn from and adapt to your car. NOT to copy paste and blindly flash. Every change made to that file was made with knowledge and/or opinions from the tuning wiki and this forum. I had never even done a stage 1 tune prior to embarking on this project. I will tell you I am running a factory bosch map sensor. If you use the search function on here you will be able to determine what pressure rating it is. I spoon fed a lot Tao, even after you insisted to communicate via email instead of through this thread as I requested you do.

Hi.
I not get it for copy paste my friend , but the error has nothing to do with your modification for e85 or injectors or...stage 1.
I don't know how make asm changes and i get your file for them. all other modification not interesting to me.
I think your map calibration grad and ofs are wrong in the last file if you use 2.5 or 3bar map sensor, but this is not important for me.
And like i told you for old file , if all asm stuff was made ok , in VCDS the actual pressure must looks ok, but is not.
Thanks for your time and support.

This is what happen after engine is after 80degree and trims begin to works. I attached a log but without modified "P" variables in file.ecu for me7logger.
The baddest thing is after i push acceleration and i left it for engine come back to idle.
I tried many dslgrade and dlsofs calculation, but all are the same.
I tried with fkkvs all 0 , and with my fkkvs from my daily file. The same thing.
With new version of Justin file , error from vcds "pressure drop between manifold and throttle' not appear again , and in group 008 , actual preassure was changed with "breake boost" .

Hi Justin. I use your file now with some mofiication for my car. It works ok with some issues and one problem!
Do you have the same issue as mine with camshaft angle when it must be changed by KFNWSE? Mine change after 5750rpm and it must be changed at 5000rpm.
The problem is from time to time when i try to make a log from 1000rpm when the boost up over 0,8 bar(approximatly) something cut and the car don;t make boost and feel like the same it is to rpm limit.
I attache the good log("log r64 3rd.csv") and KFNWSE map (WWLE is the same). The log wth cutting problem is "bad log.csv"
CDNWS is 3 and in eskonf vvt is active!
Can be in the asm code for 5120 hack an error or i must change other thing in some tables?
Thanks.

Why is wgdc still increasing when req goes over actual boost?

I am not sure you understand how the PID works if this wasn't immediately obvious

It is first log and i will adjust it in time. te problem was camshaft not change at 5000-5200 rpm conform with kfnw....

Same file only TVUB different had camshaft changed ok.
In that file too req is over actual, so i think is not a cause for my problem. Please Nyet maybe you can give me another solution for this.
I will keep actual boost and wgdc as it is maybe will increase a little to make 1.9 bar and will decrease req boost from ldr and iop.

Hi again.
In 5120 txt file with modifications you said
"KFLDRQ0(LDRQ0DY)- Divide axis by 2(Multiply by 2)
KFLDRQ1(LDRQ1DY)- Divide axis by 2(Multiply by 2)
ZDSU      - Sqrt and multiply by 2"

I found in file for these 2 maps only axis devided by 2 , but values not multiplied by 2. Does it must multiply them or not? Maybe in understood wrong your specification from upstaires.
ZDSU is not calculated. In a stock file it is 3, in justin file it is 3 too. If we apply the (sqrt(3))*2 result is 3.4641 . Did it must modified for 5120 hack or not?

Thanks.

I hoped with last night modification for some tables and with cat cleaning complete modification in my 5120 hacked file*start from justinvw file ) , resolve the problem, but nothing.
I modified to 5120 file (justin file) :
  KFMRESNL because i suspected it and i made all 0.
  CLRSHK made 5 was 0

I wrote the file and start the car. First 5 minutes car works ok at idle. After trims begin to work (STFT specially) after 2 minutes coolant temp was 70-75 degree and there idle goes crazy again. I made a log for this and I made a log for 90 degree too , where I arrived with car at home after 5 km running and stop it at idle. In 30 seconds idle goes crazy again at 90 degree coolant temp.
I checked all maps.scalars witch are modified in 5120 file from stock file and and all looks ok.
What is strange I don’t find any byte or something who tell me have other modification exclude maps, like asm changes.
With stock file modified with 4 bar map sensor car stay stable at idle with all maps/scalars same like in 5120 file only was need axis or mapss values divided or multiplied with/by 2.
Other thing who i saw in VCDS in block 8 absolute pressure with engine start is 50mbar with 5120hack file but with stock file (with the same map sensor like 5120 file) i saw 230mbar.
Some suggestion please? I don't know what happen and where is the problem.

Hi All,

Just wanted to share a stock file with a few more changes made, I got the file from page 1 of this thread.  I went through 5120 maps checklist.txt posted in this thread, as well as the page in nyet's wiki https://s4wiki.com/wiki/Manifold_air_pressure:Rescale_project

Changes I made:
ZDSU      - Sqrt and multiply by 2 (modified by blair)
PUEBKV      - Divide by 2 (modified by blair)
KMLTESG      - Divide by 2 (modified by blair)
KISRM      - Divide by 2 (modified by blair)
ZDSU      - Sqrt and multiply by 2 (modified by blair)

Things that may require more changes:
DSUGRAD      - Divide by 2 (DSUOFS was zero, otherwise divide by 2) (NOT FOUND) (fix)
NLDIAPU      - Divide axis by 2 (AXIS NOT DEFINED) (confirmed axis not defined) (fix)
PBKVVSTG   - Divide both axis and curve by 2 (NOT FOUND) (fix)


I hope to share a stock file with all the required scaling completed.  This is very close, just have to go through and find verify DSUGRAD, NLDIAPU, PBKVVSTG.

Don't use this without checking my work-- I'm still a noob and this may contain mistakes, it is untested!

Edit:

I think I found the following by comparing to an 8N0906018CB file and map pack:

dpupvdk: 1C82A
dsugrad: 1C83A
dsuofs: 1C83C

All of these are zero if I'm right.

pbkvvstg at 28A4E with the x-axis at 28A42.  

This is a stock file with 5120 asm and all the axes scaled correctly.  It is untested though!

Edit: I've recently gone over this file after many miles of testing and had some issues with pressure/load not behaving as I'd expect.  Car runs fine, but ps_w and pvdks_w don't follow closely (not a pid issue) and rl_w and rlsol do not behave as I expect

KVLAD 0x10D49 1 byte 0.007813 * X, value is currently 1.5626 (stock is 0.7813 so this is multiplied by 2) is correct, but was also incorrectly defined in my xdf as 0x10D48 (and also multiplied by 2)

I think 0x10D48 is actually KUMSRL (Conversion constant from mass flow to relative air filling).  

Read more here:
http://nefariousmotorsports.com/forum/index.php?topic=21104.0title=

Did you ever get around to testing this?

Not yet, I put my Golf in hibernation for the winter before I got around to it. I'll let you know what happens when I get around to trying this, hopefully will be this summer. 

I'd feel better about this if someone else checked my work.  I'm getting better at finding maps but I'm still new to this.

Best,
Blair

I went over your file and everything looked correct, so i went ahead and modded it to my suit and flashed it to my car.
And everything looks to be working perfect. Thanks so much for the file  ;D

That's awesome, really happy to hear it worked!

This file is working great on my car as well.  Tested today, can post logs if anyone is interested.

Do it.

Here you go!


edit: I'm working on an *.ecu file too I'll post that when it's done and somewhat tested

Many thanks to nyet for writing the perl script for this.

In your files asm divisions are made from 4D 65    to A7 32 not to A6 32
must be like you or 4D 65 to A6 32 ?

Updated files (completely untested, not checksummed)

KUMSRL reverted to stock as I believe it was incorrectly modified

Dude you and your divisions, stop spamming the forum and get a clue.

Do you really think it's going to make any difference dividing by 12966 or 12967? A whopping difference of 0.001%
Get over it.

25933/2 is 12966.5 you need to round one way. Some people rounded one way, others did it the other way.
Does it make any difference? no.

I tried the 5120 asm for the file that i currently use (06a906032HJ).

Can someone have a look at it and confirm if it's done correct?
I will attach the bin here.